refine qwen3_vl_moe experts forward#1413
Merged
WeiweiZhang1 merged 3 commits intomainfrom Feb 6, 2026
Merged
Conversation
Signed-off-by: WeiweiZhang1 <weiwei1.zhang@intel.com>
Contributor
There was a problem hiding this comment.
Pull request overview
This PR refines the expert forward pass logic in the Qwen3 VL MoE (Mixture of Experts) implementation, optimizing how experts are selected and invoked during inference.
Changes:
- Wrapped expert mask computation in
torch.no_grad()to prevent gradient tracking for this operation - Restructured the expert iteration logic to skip unused experts when
calibrate_all_expertsis False - Simplified tensor indexing by removing unnecessary
.squeeze(0)operation
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
wenhuach21
approved these changes
Feb 6, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
Please briefly describe your main changes, the motivation.
Type of Change
Related Issues
Fixes or relates to #
Checklist Before Submitting
The measured accuracy of the quantized model are the same as before.